The Natural Language Processing and Information Retrieval Group at UNED currently consists of twenty researches (9 doctors, permanent academic staff), from UNED's Escuela Técnica Superior de Ingeniería Informática. The group began its research activities in 1993 and, since then, it has been involved in 14 research projects funded by public national and European programmes. In the last five years, more than one hundred publications have been generated in high impact journals and conferences, 8 PhD theses have been obtained and 10 are ongoing.
The research line of interest for the group are:
The members of NLP&IR-UNED also participate in the organization of international evaluation campaigns and in the proposal of new evaluation frameworks within the Cross Language Evaluation Forum (CLEF) and other communities: iCLEF, CLEF QA, AVE, Web People Search.
The most relevant projects (see http://nlp.uned.es/projects.html) for this proposal in which NLP&IR-UNED has been involved are: QEAVis: Quantitative Evaluation of Academic Websites Visibility (TIN 2007-67581-C02-01) in which HLT are applied to measure the visibility of academic web sites, focusing on Humanities and the presence of the Spanish language; TrebleCLEF: Evaluation Best Practice and Collaboration for Multilingual Information Access (FP7, ICT-1-4-1 #215231) devoted to evaluations of multilingual information access systems; TEXT-MESS: Minería de Textos Inteligente, Interactiva y Multilingüe basada en Tecnología del Lenguaje Humano (TIN2006-15265-C06 ), focused in the development of intelligent, multilingual and interactive technologies for text mining, image and semi-structured documents search; MedIEQ: Quality Labeling of Medical Web content using Multilingual Information Extraction (FP6, Public Health A/790604) whose goal was the development of an automatic labelling application in the health domain; MultiMATCH: Multilingual/Multimedia Access to Cultural Heritage (FP6-2005-IST-5 #033104) focused on the information access systems in the European cultural heritage domain; and R2D2: Recuperación de Respuestas en Documentos Digitalizados (TIC2003-07158-C04), focused in the evaluation of cross-lingual question answering systems.